The aim of this study was to examine whether the preterm birth (PTB) risks according to maternal age is altered by a woman's marital status and chronic stress among non-Hispanic (N-H) White, N-H Black, Hispanic, and Asian women. This researcher analyzed the Pregnancy Risk Assessment Monitoring System data for New York City and Washington State linked with the birth certificates for 2004–2007. The sample included 6344 singleton live births without birth defects to women aged 18 years or older identified as N-H White, N-H Black, Hispanic, or Asian. The outcome was PTB. Maternal age-specific PTB rates were calculated according to race/ethnicity, marital status, and chronic stress. Linear trends of PTB rates with maternal age were evaluated by the Mantel–Haenszel χ2 test. Marriage had a protective effect against PTB at advancing maternal age across racial/ethnic groups. The health benefit of marriage was strong, particularly among the married N-H Black and Asian women, manifested as a maternal age-related decrease in the PTB rate (reverse-weathering). In contrast, women not married showed a maternal age-related increase in the PTB rate (weathering) across the racial/ethnic groups. Under higher chronic stress, married women generally experienced less weathering about PTB. These patterns were observed with noticeable racial/ethnic variations. Acknowledging the different dynamics among maternal age, marital status, and chronic stress by race/ethnicity could help shed light on the psychosocial mechanisms underlying the racial/ethnic inequalities in PTB in the United States. To that end, future studies should use more nuanced measurements of paternal support and chronic stress.